cobem-2017-1243 design of active suspension mechanism

24th ABCM International Congress of Mechanical Engineering December 3-8, 2017, Curitiba, PR, Brazil

COBEM-2017-1243

DESIGN OF ACTIVE SUSPENSION MECHANISM

João Victor Dini Colatto

Brave-Brasil Electric Vehicles, R. Osasco, 268, Cajamar – SP, CEP 07753-040

[email protected]

Rafael Cipelli Pellicci

Maua Institute of Technology, Praça Mauá, 1, São Caetano do Sul – SP, CEP 09580-900

[email protected]

Mateus Fellin Macedo


[email protected]

Fernando Teixeira Monteiro


[email protected]

Fernando Malvezzi


[email protected]

Abstract. This work deals with the design of a 3-DOF active suspension mechanism that is able to control Vehicle

Body Roll, Toe and Camber angles. This type of mechanism is able to improve handling performance and vehicle

stability control, when it is compared to a passive suspension. Several mechanism designs were generated and one of

them was described in this work. Firstly, the initial design of a product was chosen for all the components of the new

suspension system. It includes the layout for the mechanism to be assembled on a vehicle. Then, using the commercial

software ANSYS, a structural analysis was developed by applying the finite elements method. The input forces

considered in these analyses correspond to the maximum values obtained considering a vehicle executing two different

maneuvers. The first one is the fishhook maneuver, which was used to analyze the vehicle rollover tendency. The

second one, a breaking test, was applied to analyze the vehicle behavior during hard breaking conditions. The results

show that the mechanism stresses caused by the forces imposed on it during these maneuvers are below the allowable

stress.

Keywords: Suspension Mechanism, Active Suspension System, Finite Element Analysis, Vehicle Dynamics.

1. INTRODUCTION

Traffic accidents have generated a large number of deaths in Brazil. In 2013, the DPVAT insurance paid 54,880

death benefits and 444,000 invalidity benefits (VIAS SEGURAS, 2014). Among these accidents, 18% of them were

caused by car skidding and 3.3% were caused by rollover. As an attempt to attenuate this scenario, stability control

systems for passenger vehicles have been employed. The most relevant control measurement applied for stability

control systems is the ESP – Electronic Stability Program Stability based on active brake control. Several international

studies carried by car manufacturers and safety agencies show that the ESP system can reduce by up to 80% the

accidents caused by skidding, which is one of the main causes of serious traffic accidents. In Europe, the use of ESP in

all vehicles resulted in an annual reduction of 4,000 fatalities and 100,000 injuries (BOSCH, 2017).

Although the ESP can contribute for vehicle stability control, this system presents some disadvantages, such as the

vehicle speed reduction (Eduardo, 2009; Sohn and Park, 2012). In contrast, other means of control, so as active

suspension and active steering systems (Deur, et al, 2011) can contribute to vehicle stability control without the loss of

velocity unlike ESP. Moreover, many studies have shown that currently there is a tendency to integrate two or more

technologies for stability control, such as in (Lu, et al; Rengaraj and Crolla, 2011).

J. V. D. Colatto, R. C. Pellicci, M. F. Macedo, F. T. Monteiro and F. Malvezzi Design of Active Suspension Mechanism

Malvezzi and Hess-Coelho proposed a mechanism that controls the camber and rear steering angles, as well as

provides an anti-roll control. This mechanism was designed on the base of a parallel kinematic structure that can

contribute to vehicle stability, as shown in (Malvezzi, 2014; Malvezzi and Hess-Coelho, 2015).

This work aimed to design an active suspension mechanism, based on the mechanism presented in (Malvezzi,

2014). Firstly, the initial design was chosen, including all the components of the new suspension, as well as an assembly

viability study. Following, a structural analysis was conducted to assure that every part complies with structural

demands. Finally, the designs of all mechanism parts were primed.

2. COMPUTACIONAL PROCEDURE

In this work a design for a 3-DOF active suspension mechanism was developed. It is based on the RTC mechanism

(a mechanism that can actuate on Vehicle Body Roll, Toe and Camber angles) proposed by Malvezzi (2014). According

to the results presented in that work, the vehicle equipped with RTC active mechanism had an improved handling

performance and stability control when compared to the original one, which is equipped with a passive suspension.

For the advantages of applying the RTC mechanism to make able, a layout study had to be conducted, so the

assembly could be viable in a vehicle. To achieve this goal, the flowchart presented in Fig. 1 was employed. The

commercial software ANSYS was used for structural analysis.

Figure 1. Flowchart for design of the active suspension mechanism

Figure 2-a shows the RTC mechanism proposed by Malvezzi (2014). Figure 2-b presented the first mechanism

design generated in this work and Fig. 2-c shows the main parts of the final mechanism design of this work.

Figure 2. (a) RTC Mechanism; (b) First mechanism design; (c) Final mechanism design

(a) (b) (c)


According to Fig. 2-b, the first active motion sets the camber angle, which is achieved by the action simultaneously

of the actuators 2 and 3 in the same direction. The second active motion sets the toe angle, obtained by the movement of

the actuators 2 and 3 in the opposite direction. The third motion can be either passive or active. In the passive motion, it

is assumed that the car does not perform evasive maneuvers or cornering. In this case, the actuator 1 does not exert

force on the sprung mass (so there is no energy consumption) and the wheel travel is influenced only by the action of

the spring and the shock absorber. On the other hand, the active motion provides anti-roll control by a force exerted by

the actuator 1 on the sprung mass (body).

In order to analyze the mechanism mobility the Gruebler-Kutzbach criterion (Tsai, 1999), defined by Eq. (1), is

applied.

� � �� 1 ∑� (1)

where M is the mobility of the mechanism, the index λ corresponds to the space dimension where the mechanism is

supposed to function, n is the number of links in the mechanism (including the fixed base), j is the number of joints in

the mechanism, ff is the number of degrees of freedom of relative motion permitted by joint f.

According to Figure 3, the mechanism has 8 limbs and 9 joints. By applying the Eq. (1), the mobility is equal to 3:

� � 6�8 � 9 � 1 15 � 3 (2)

Figures 3-a and 3-b show the mechanism parts identification used to apply Gruebler-Kutzbach criterion. In Fig. 3-c

the active joint R is equivalent to the spring-damper-actuator 1 assembly.

Figure 3. (a) and (b) Mechanism parts identification; (c) Mechanism graph representation

3. PREPROCESSING FOR FINITE ELEMENT ANALYSIS

All the required data for the simulations were extracted from the work of Malvezzi (2014). These data were

generated from simulating the vehicle dynamic behavior on the CarSim software. It was employed a C-Class vehicle

which 1,274 kg, available in the CarSim data library.

Figure 4. Fishhook Maneuver: (a) Test Procedure; (b) Steering wheel input

(a) (b)

(a) (b) (c)

Joints legend

P: Prismatic (active)

R: Revolute (active)

R: Revolute

S: Spherical

U: Universal


The input forces in the finite elements analysis correspond to the maximum values obtained considering a vehicle

executing two different maneuvers. The first one is the fishhook maneuver, which is used to analyze the vehicle rollover

tendency. This maneuver consists of applying an angular turn to the steering wheel to remove the vehicle from linear

movement. After that, a fast correction response is applied and the steering wheel is turned in the opposite direction.

The peaks of forces and moments on the rear left wheel were achieved in the second part of the fishhook maneuver

(after overcorrection in Fig. 4).

The second maneuver considered the vehicle during hard braking conditions. The weight supported by the rear left

suspension system is 2,000N. In addition, it was adopted a coefficient of static friction of 0.8 between the tire and the

pavement.

Table 1 shows such values and net configurations employed in the finite element analysis. It is important to note that

the tire forces and moments were defined considering the CarSim reference axis (Fig. 2-a), which is different from the

ANSYS reference axis (Fig. 5).

Table 1. Inputs data employed in the finite element analysis (rear left wheel).

Tire Forces and

Moments (CarSim

reference axis)

Braking

Test

Fishhook

Maneuver Units

Fx 1,600 -9 N

Fy 0 3,400 N

Fz 2,000 5,015 N

Mx 0 -115 N.m

My 0 0 N.m

Mz 0 53 N.m

Element Types Hex Dominat Hex Dominat -

Midside Nodes Dropped Dropped -

Element size range 1 to 10 1 to 10 mm

The lower control arm is fixed to the chassis by means of bushes. Therefore, the boundary condition used was a

cylindrical support with free tangential movement. The upper control arms and their actuators were represented by

springs with a rigidity of 2,000N/mm. Only the actuator 1 along with its shock system was represented by a spring’s

rigidity of 6,500N/mm (Fig. 5).

Figure 5. Pre-processing: mechanism CAD model


The parts that do not have any movement between than were bonded into a single component. As a result, the

suspension system was composed only by three more important parts: the spindle, the lower control arm and the joint

linking the lower control arm and the spindle (Fig. 5). The final geometry is shown in Fig. 6 and 7.

Figure 6. Frictionless contact

Figure 7. Spherical joint

4. SIMULATIONS AND RESULTS

As the results from the simulations were non-linear, the ‘large deflection’ command was enabled. As a result, the

Ansys software executes a series of interactions adding portions of forces (steps) until the convergence is reached.

Figures 8 and 9 show such interactions for one simulation loop.


Figure 8. Iteration process for braking test

Figure 9. Iteration process for fishhook maneuver

4.1 Simulation Loops

By the end of the simulation 21 loops were done for each maneuver. The suspension’s geometry was modified in

each one these loops in order to minimize stress concentration regions. Some of these modifications were, for instance,

the addition of fillets to eliminate sharp edges, ribs to increase torsion rigidity and minimize over-dimensioned

components. Fig. 10, 11 and 12 portrait the most relevant modification loops (loops 4, 7 and 11).

In the loop 4 it was noticed a geometrical singularity, meaning that for both simulations with different loads there

was a stress concentration point. It was solved by adding a thicker fillet in that point (Fig. 10).

In loop 7 it was noticed that the stresses for both braking and fishhook maneuvers decreased and are located in

different regions of the suspension. During the former, the highest stress is located in the interior part of the lower

control arm, close to the connection point between the chassis and the control arm. It was solved by adding ribs and

then closing the control arm’s profile (Fig. 11).


Figure 10. Loop 4 - Results for: (a) and (b) braking test; (c) and (d) fishhook maneuver

Figure 11. Loop 7 - Results for: (a) and (b) braking test; (c) and (d) fishhook maneuver

(a) (b)

(a)

(b)

(c) (d)

(c) (d)


It was noted in loop 11 that for both maneuvers the stresses decreased and again they changed place. For braking,

the maximum stress was concentrated in the crimping region of the pin in the articulation of the control arm, whereas in

the fishhook the fillet of 3 mm was not sufficient to reduce the stress, since it only changed from one side to the other

side of the spindle (Fig. 12). The obtained results in the loop 11 are not realistic, because after mesh-refinement in

subsequent loops the equivalent stress increased to 650 MPa.

Figure 12. Loop 11: Results for: (a) and (b) braking test; (c) and (d) fishhook maneuver

4.2 Final Results

During each simulation, it was possible to observe loop the reduction and change of position of the maximum Von-

Misses Stress. However, the initial target for the project was to obtain equivalent stress lower than 400 MPa. To achieve

this target, it was necessary to decrease the diameter of the pin, thus, the fixing nut. This allowed for remodeling the

spindle and removing the cut that allowed the mechanical assembly of the suspension. This change resulted in a uniform

surface reducing the stress to 353.17 MPa in the fishhook maneuver.

At the critical instant of the fishhook maneuver (position 2 in Fig. 4-a), the vehicle tends to roll out of the turn, with

the body approaching the left rear tire, compressing the spring, trailing the upper control arm and causing a buildup of

stress in the fillet of the spindle, as shown in figure 13-b.

(d)

(a) (b)

(c)


Figure 13. Final results of fishhook maneuver

After several loops for the braking maneuver, by changing the geometry, the point of stress concentration did not

change, remaining in the pin-set region of the pin with the joint. It was possible to reduce the maximum stress to 412.57

MPa by increasing thickness of the joint. In the braking situation, with the proposed geometry, the forces and moments

are transferred from the tire-pavement contact to the lower control arm. It causes a bending stress in the pin, according

to Fig. 14-b.

Figure 14. Final results of braking test

5. CONCLUSIONS

This paper describes the design of a 3-DOF active suspension mechanism that is able to control the Toe, Camber

and Body Roll angles simultaneously. Firstly, the study of the layout for the mechanism assembly on a vehicle was

carried out. Second, using the commercial software ANSYS, a structural analysis was developed by applying the finite

elements method. In this analysis the forces and moments in the tire-road contact area were obtained making use of a C-

Class vehicle running in two different situations: in the fishhook maneuver and under hard braking condition. Finally,

the obtained results were analyzed. They revealed that the maximum equivalent stress is 2.7 times lower than the

ultimate stress of high tensile steel. In order to build a mechanism prototype to be assembled on a vehicle, in the future

woks a topology optimization, a fatigue and an impact analyses will be conducted. This vehicle will be used only to

evaluate the performance of this mechanism at a proving ground.

6. ACKNOWLEDGEMENTS

The authors would like to acknowledge the Brave-Brasil Electric Vehicles for supporting and facilitating the

research of this work.

(b) (a)

(b) (a)


7. REFERENCES

Vias Seguras, 2014. Brazilian traffic accidents information. 24 nov. 2014 <www.vias-seguras.com>.

BOSCH Automotive Technologie. Informations about ESP – Eletronic Stability Control. 17 jun. 2017

<www.bosch-tecnologiaautomotiva.com.br>.

Eduardo, G. P., 2009. Optmal neurocontroller by genetic algorithms for multiple vehicle dynamics active systems at

yaw. Ph. D. Thesis, University of São Paulo, São Carlos.

Sohn, J. e Park, S., 2012. “Effects of camber angle control of front suspension on vehicle dynamics behaviors”,

Journal of Mechanical Science and Technology, pp. 307-313.

Deur, Z.; Hancock, M.; Assadian, F., 2011. “Design and comparative study of yaw rate systems with various

actuators”. In: SAE International.

Lu, S. B. et al., 2011. “Integrated control on MR vehicle suspension system associated with braking and steering

control”, Vehicle System Dynamics, V. 49, pp. 361-380.

Rengaraj, C.; Crolla, D., 2011. “Integrated chassis control to improve vehicle handling dynamics performance”. In:

SAE International.

Malvezzi, F., 2014. New Approaches to the development of vehicle suspension: the utilization of parallel

mechanism.Ph.D. thesis, University of Sao Paulo, Sao Paulo. In Portuguese.

Malvezzi, F.; Coelho, T. A. H., 2015. “Modeling, feasibility and performance analyses of a 3-DOF parallel

mechanism employed in a rear vehicle suspension”, International Journal of Vehicle Systems Modelling and Testing,

V.10, No 1 pp. 53-73. Tsai, L.-W., 1999. Robot analysis: the mechanics of serial and parallel manipulators. New York: John Wiley & Sons.

8. RESPONSIBILITY NOTICE

The author(s) is (are) the only responsible for the printed material included in this paper.

cobem-2017-1243 design of active suspension mechanism

Documents